Functional Robustness on MobileSafetyBench Low-Risk N=35 1.0
94.3Task Success RateAura
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Aura2026.02 | 94.3 | 68.54 | |
| Doubao-ProMode=Reasoning-Heavy2026.02 | 77.1 | 513 | |
| Doubao-StandardMode=Balanced2026.02 | 74.3 | 441 |
| Method | Links | ||
|---|---|---|---|
| Aura2026.02 | 94.3 | 68.54 | |
| Doubao-ProMode=Reasoning-Heavy2026.02 | 77.1 | 513 | |
| Doubao-StandardMode=Balanced2026.02 | 74.3 | 441 |